The human neuroblastoma SH-SY5Y cell line as a model to assess β-amyloid neurotoxicity: A systematic review and meta-analysis

The SH-SY5Y human neuroblastoma cell line is widely used as an in vitro model of β-amyloid (Aβ) neurotoxicity in Alzheimer’s disease (AD). However, the lack of standardized protocols for assessing Aβ toxicity - including differentiation strategies for SH-SY5Y cells - limits the comparability of results across studies. To address these issues, we conducted a systematic review and meta-analysis to evaluate how methodological factors influence Aβ-induced toxicity in SH-SY5Y cells. We included 359 eligible studies encompassing 1,192 MTT-based comparisons of cell viability between Aβ-treated and control SH-SY5Y cells. A three-level meta-analysis estimated mean cell viability after Aβ exposure at 63% of control levels (95% CI [61.6; 64.3]), with very high heterogeneity (I²=99.6%). Meta-regression identified significant associations between increased toxicity and higher Aβ concentrations, longer exposure durations, and the use of peptide preparations described as fibrils. Conversely, differentiation protocols, duration, and cell density did not significantly influence toxicity outcomes. Reporting quality was often poor, with frequent omissions regarding cell line origin, authentication, contamination testing, Aβ preparation details and nature of the experimental unit. Overall, our findings show robust Aβ toxicity in SH-SY5Y cells, primarily driven by dose, exposure time, and Aβ aggregation state, but not cell differentiation status. Our conclusions highlight the critical need for better reporting of Aβ exposure parameters to enhance reproducibility and translational potential in AD research.